Project Summary: Surgical resection remains the mainstay of the treatment of neurological malignancies. Blood oxygen level dependent functional MRI (BOLD fMRI) can identify essential functional areas of the brain (eloquent cortices) adjacent to brain tumors. BOLD fMRI does not directly detect neuronal activity. Rather, it measures the vascular response that results from increased neuronal activity (neurovascular coupling). It is known that tumor neovasculature reacts to stimuli in a diminished manner. Also, the blood vessels may already be maximally dilated in the setting of a hypoxic malignant tumor. Thus, due to the muted vascular response, the contrast necessary for a reliable fMRI signal appears to be compromised. Therefore, BOLD fMRI appears to be less accurate in the most important location for neurosurgical treatment planning ? the area directly adjacent to the tumor. We hypothesize that the location of an abnormal BOLD response (false negative) adjacent to malignant brain tumors can be identified by MRI measurements of abnormal tumor neovasculature, defined by routine MRI data (including perfusion) and vascular reactivity maps, quantified by breath-holding MRI (BH MRI). We further hypothesize that the accuracy of BOLD fMRI in the vicinity of brain tumors can be improved by calibrating the BOLD response by incorporation of BH MRI. The innovation of this grant is to incorporate an independent measurement of cerebrovascular reactivity (BH MRI) into the BOLD fMRI analysis to overcome the false negative activation due to neurovascular uncoupling by tumor neovasculature. Specific Aims: SA1. Determine the relationship between fMRI accuracy and the presence of abnormal tumor neovasculature as defined by anatomical MRI parameters and measurements of vascular reactivity (BH MRI). SA2. Use calibration of the BOLD fMRI data by measurements of vascular reactivity (BH) to compensate for neurovascular uncoupling and increase the accuracy of BOLD fMRI adjacent to tumors. SA3. Determine when independent measurements of vascular reactivity can provide additional information and should be performed to improve the accuracy of BOLD fMRI. Expected outcomes: 1. The accuracy of BOLD fMRI will decrease in the presence of abnormal tumor neovasculature that affects the brain parenchyma. 2. Incorporation of BH MRI into the BOLD fMRI calculations will significantly improve the accuracy of identifying eloquent cortices adjacent to brain tumors with abnormal neovasculature. 3. Patients with tumors which exhibit abnormal parenchyma brain enhancement and perfusion will benefit from incorporation of BH MR into BOLD fMRI calculations. Health relatedness: improve the identification of eloquent cortices adjacent to brain tumors by fMRI, which will have an immediate, positive impact on brain tumor resections. The broad, long-term objectives of the current proposal is not limited to brain tumors, but will extend to other conditions, which affect vascular reactivity and consequently, the accuracy of fMRI, including normal aging and pathological conditions such as arteriovenous malformations, strokes, vascular stenosis, or microvascular disease.